Web applications that scale to a single organization

Unlike Project Specific Websites, the architecture of Enterprise Extranets allows one organizational container to hold multiple projects. This eliminates the redundancy of master lists such as contacts, phone numbers, cost codes and user security profiles. With one username and password, a user can access (security permitting) all of the projects within an organization.

Enterprise Extranets are not only designed for, but provide the most benefit to, those enterprises that track every project from it. The key benefit is that an organization has a single source of project information enterprise-wide, providing efficiencies in the form of common access and format across projects as well as the ability to aggregate data between projects (i.e. total dollar volume of retail vs. healthcare or, workload statistics across all project managers).

While the structure of the Enterprise Extranet provides greater efficiency for those organizations hosting projects, other team members (those collaborating on projects hosted by others) are left with the same integration problems that hosting organizations face with Project Specific Websites. That is, data on some of their projects resides on one system while data from other projects resides on another system.

Furthermore, in both the Project Specific Website and Enterprise Extranet, the hosting organization owns and controls all of the data for a project. This leaves non-hosting organizations at the mercy of the host should it be decided to archive the data and remove it from the site, or to limit or revoke the security rights of users outside the host's organization. 

Most importantly, however, being organization centric, the Enterprise Extranet carries an inherent flaw in efficient collaboration in that one organization acts as a public clearinghouse for all information on a project. In reality, organizations share information in a complex network of privileges with other organizations. For instance, given all of the information from one organization, a portion will be made available to all other organizations while another portion will be made available to a select few. Yet another portion will be made available to another select few, but a different select few than the previous portion. Finally, the remaining portion is not shared with any other organizations, but tracked for internal use only. The diagrams above illustrate the contrast between communications in Enterprise Extranets and the real world.

In an Enterprise Extranet, the real world level of control over the distribution of information is only possible for the hosting organization. Technically, the hosting organization may be able to employ the security features of its application to create these complex gateways between other organizations, but it would be awkward at best and incomplete at worst. For example, if one non-hosting organization wanted to share information with another non-hosting organization but not with the hosting organization, it would have to trust the hosting organization to essentially not look at the information. Even here, while the honest hosting organization is not privy to the content of the information, it still knows the type of information being shared and that it is not being shared with it.

Enterprise Extranets are built on the same architecture as client-server applications in which each organization is isolated in the shell of a local area network (LAN). Because there is generally only one organization on a LAN, the functionality for multiple organizations to share and intertwine process on their own terms is not useful. So by porting client-server models, as is, to the Internet, Enterprise Extranets have greatly extended the reach of LANs but, unfortunately, have kept them every bit as discrete as they were before the Internet. As a result, these applications have not even scratched the surface of the value of a large shared network. To the contrary, they only magnify the inefficiency of many private networks.

Enterprise Extranets revolve around a single organization that contains projects. When deployed, on a LAN or the Internet, they effectively return value only to the account-holding organization. These applications assume that several organizations revolve around one organization that contains a project. The reality is, organizations revolve around projects, not other organizations. More important, projects are not contained by organizations.

Ironically, Enterprise Extranets are sold as solutions that connect organizations. In reality, they separate organizations. Fully deployed in a supply chain, they act as little more than highly efficient fax machines. They provide value to hosting organizations, but when used on every project by every general contractor, the net value returned to any single project, and thus the owner, will be negative.